基于fpga的双目视觉图像预处理系统设计与实现-计算机技术专业论文.docxVIP

研究生优秀毕业论文 Abstract Abstract 硕上学位论文 Abstract With the development of the theory and technology of computer vision，binocular vision is gradually becoming an important part of computer visio．Many universities and graduate schools in the field has done a lot of research and achieved considerable achievements．Binocular vision technology has a broad application prospects in aerospace， mobile robot，industrial detection，intelligent transportation and other fields．With the increasing complexity of binocular vision applications，the preferments of binocular vision system based on traditional embedded equipment are more and more difficult to meet the actual needs． In the image preprocessing stage of traditional binocular vision system，using software control general CPU to preprocessing images is cost large．In order to improve the efficiency of image preproeessing stage，usually by improve the algorithm or the use of more efficient hardware platform．This paper from the start with the efficient hardware platform，making parallel design by FPGA for image preprocessing stage to reduce the execution time，and improve overall efficiency of binocular vision system based on embedded equipments． This system using the ZYNQ series chip，that is FPGA+ARM structure chip．The system used Verilog hardware description language to describe the part of hardware on FPGA．and used C language to complete the development of software on ARM．The FPGA in this chip used to optimize the time—consuming algorithm The ARM in this chip used to complete the control function of the system，and achieve the algorithm that use FPGA is difficult to optimize． The main work of this paper： According to the characteristics of ZYNQ，and using the method of hardware—software CO—design to divide hardware and software fu